2019/11/16 by Morteza Raeisi, Raeisi, Morteza, Florent Bonneu +3
Economics, Econometrics and Finance · Mathematics · #60G55 #62H30 #62M30 #62P12 #Applications (stat.AP) #Artificial intelligence #Cartography #Computer science #Data mining #Econometrics #FOS: Computer and information sciences #Geography #Inference #Mathematics #Methodology (stat.ME) #Point (geometry) #Point process #Point processes and geometric inequalities #Process (computing) #Scale (ratio) #Spatial and Panel Data Analysis #Statistics
paper · pdf · doi:10.48550/arxiv.1911.06999
openalex publication_date 2019/11/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Because most natural phenomena exhibit dependence at multiple scales like\nlocations of earthquakes or forest fire occurrences, spatio-temporal\nsingle-scale point process models are unrealistic in many applications. This\nmotivates us to construct generalizations of classical Gibbs models. In this\npaper, we extend the Geyer saturation point process model to the\nspatio-temporal multi-scale framework. The simulation process is carried out\nthrough a birth-death Metropolis-Hastings algorithm. In a simulation study, we\ncompare two common methods for statistical inference in Gibbs models: the\npseudo-likelihood and logistic likelihood approaches that we tailor to this\nmodel. Finally, we illustrate this new model on forest fire occurrences\nmodeling in Southern France.\n